SMART WARS: KardiaBand Bests Apple Watch Series 4 for Detecting Arrhythmia

Results of the SMART WARS study suggests KardiaBand devices outperformed the Apple Watch Series 4 for accuracy and sensitivity for detecting atrial fibrillation.

A head-to-head comparison of the automated algorithms used with each device in an outpatient setting, results of the study suggest the automated algorithm used in KardiaBand devices provided greater accuracy and sensitivity for detecting paroxysmal arrhythmias compared to the Apple Watch Series 4, but investigators note results also provide evidence in support of clinician assessment to improve accuracy of single-lead electrocardiogram devices.

“Our study provides real-world validation of commercially available smart watch algorithms for the detection of AF in an elderly, ambulatory population. The KardiaBand was more accurate at ascertaining heart rhythm and detecting AF than the Apple Watch Series 4. The overall accuracy of the Apple Watch Series 4 ECG classification considering all rhythm strips was substantially lower (66%) than reported by Apple (88%),” wrote investigators.

The utility of smartwatches and portable ECG devices for detecting arrhythmias has gone from science fiction to reality in recent decades. With the capability and popularity of wearable devices growing in stride with one another, optimization of leveraging these technologies to improve identification and, as a result, reduce the burden of arrhythmias has taken centerstage for many electrophysiologists. Designed by investigators from the Department of Cardiology at Monash University conducted the current study with the intent of comparing the diagnostic utility of the KardiaBand and Apple Watch Series 4 devices.

To do so, investigators designed their study to leverage data from consecutive recordings taken from patients attending cardiology outpatient clinics with both devices concurrently. Both automated diagnoses and blinded single-lead electrocardiographic tracing interpretations from 2 cardiologists were analyzed and additional analyses were conducted to determine the effect of the combined device and clinical interpretation.

In total, 125 patients were recruited from December 2019-March 2020 for participation in the study. Inclusion criteria for the study included being 65 years of age or greater and the ability to tolerate lying supine for 12-lead electrocardiography. Investigator pointed out 65 years was chosen as the age cutoff because of the increased prevalence of atrial fibrillation in older patients. The 125-patient cohort had a mean age of 76±7 years and 62% were men.

Upon analysis, results suggested the accuracy of the automated rhythm assessment algorithm was greater with the KardiaBand device than with the Apple Watch Series 4 device, with observed accuracies of 74% and 65%, respectively. When assessing sensitivity and negative predictive value for detection of atrial fibrillation, the KardiaBand device returned values of 89% and 97%, respectively. For the Apple Watch Series 4, assessments of sensitivity and negative predictive value for detection of atrial fibrillation yielded values of 19% and 82%, respectively. In models using hybrid automated and clinician interpretation, the overall accuracy of the KardiaBand device improved to 91% and the accuracy of the Apple Watch Series 4 device increased to 87%.

“These findings suggest that although these devices’ tracings are of sufficient quality, automated diagnosis alone is not sufficient for making clinical decisions about atrial fibrillation diagnosis and management,” investigators wrote.

This study, “Comparison of 2 Smart Watch Algorithms for Detection of Atrial Fibrillation and the Benefit of Clinician Interpretation: SMART WARS Study,” was published in JACC: Clinical Electrophysiology.